Over 60 million Americans use herbal medicines, of whom one fourth also take prescription drugs. Physicians often are unaware of herbal use and of possible drug/herb interactions. Ginseng and ginkgo, enhancers of physical and mental performance, are two of the most widely taken herbals. We propose a double- blind, randomized, prospective study of effects of ginseng and ginkgo on 1) disposition of probe drugs, 2) cognitive function, and 3) glutathione-S-transferase (GST) and quinone reductase (NQO1), enzymes implicated in chemoprevention of cancer. Probe drugs will be administered to study effects of herbs on their disposition, not for therapeutic effect. Ideal probes must be safe, well tolerated, have minimal pharmacological effect, and share known metabolic pathways with other clinically used drugs. Medically stable drug-free non-smokers will be enrolled. During a 4-week single-blind run-in subjects will be given a 4-drug probe cocktail: caffeine to study cytochrome P4501A2 (CYP1A2), dextromethorphan for CYP2D6, buspirone (and endogenous cortisol) for CYP3A and fexofenadine for P-glycoprotein. Losartan will be given separately for CYP2C9. These enzymes metabolize over 95 percent of clinically used drugs. Enzyme activities will be determined by assaying appropriate blood and urine specimens for probe drugs and metabolites. Cognitive function will be tested and blood lymphocytes collected for measuring GST and NQO1 activities. Sixty subjects will then be randomly assigned to one of 4 double-blind treatment groups of 15 each: 1) ginseng extract (Ginsana), 2) ginkgo extract (EGb761), 3) both herbs, or 4) matching placebos. Tolerability of herbs will be determined. After 6 to 8 weeks of twice daily treatment with study agents, all effect parameters will be reevaluated: probe drug pharmacokinetics, cognitive function, and GST and NQO1 in blood lymphocytes. Interactions of chronic ginseng and ginkgo with drug-metabolizing pathways and with cognitive function will thus be determined.